conftrace_

2022 SemEval SemEval 2022

RIGA at SemEval-2022 Task 1: Scaling Recurrent Neural Networks for CODWOE Dictionary Modeling

Abstract

AbstractDescribed are our two entries “emukans” and “guntis” for the definition modeling track of CODWOE SemEval-2022 Task 1. Our approach is based on careful scaling of a GRU recurrent neural network, which exhibits double descent of errors, corresponding to significant improvements also per human judgement. Our results are in the middle of the ranking table per official automatic metrics.

Authors

Eduards Mukans, Gus Strazds, Guntis Barzdins

Topics

Machine Learning > Core Methods > Representation Learning Machine Learning > Optimization & Theory > Neural Network Optimization Deep Learning > Architectures > Neural Networks

Keywords

representation learning double descent neural network optimization recurrent neural network dictionary modeling

More from SemEval 2022

All SemEval 2022 papers →

TLDR at SemEval-2022 Task 1: Using Transformers to Learn Dictionaries and Representations

MMG at SemEval-2022 Task 1: A Reverse Dictionary approach based on a review of the dataset from a lexicographic perspective

Edinburgh at SemEval-2022 Task 1: Jointly Fishing for Word Embeddings and Definitions

Uppsala University at SemEval-2022 Task 1: Can Foreign Entries Enhance an English Reverse Dictionary?

BL.Research at SemEval-2022 Task 1: Deep networks for Reverse Dictionary using embeddings and LSTM autoencoders

JSI at SemEval-2022 Task 1: CODWOE - Reverse Dictionary: Monolingual and cross-lingual approaches